Simple and Effective Multi-Paragraph Reading Comprehension

  title={Simple and Effective Multi-Paragraph Reading Comprehension},
  author={Christopher Clark and Matt Gardner},
  • Christopher Clark, Matt Gardner
  • Published 2018
  • Computer Science
  • ArXiv
  • We consider the problem of adapting neural paragraph-level question answering models to the case where entire documents are given as input. [...] Key Method We sample multiple paragraphs from the documents during training, and use a shared-normalization training objective that encourages the model to produce globally correct output. We combine this method with a state-of-the-art pipeline for training models on document QA data. Experiments demonstrate strong performance on several document QA datasets. Overall…Expand Abstract

    Figures, Tables, and Topics from this paper.

    BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
    • 9,745
    • Open Access
    Deep contextualized word representations
    • 4,082
    • Open Access
    QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension
    • 438
    • Open Access
    MS MARCO: A Human Generated MAchine Reading COmprehension Dataset
    • 501
    • Open Access
    Know What You Don't Know: Unanswerable Questions for SQuAD
    • 581
    • Open Access
    HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering
    • 270
    • Open Access
    QuAC : Question Answering in Context
    • 188
    • Open Access
    Passage Re-ranking with BERT
    • 170
    • Open Access
    AllenNLP: A Deep Semantic Natural Language Processing Platform
    • 463
    • Open Access


    Publications referenced by this paper.
    Glove: Global Vectors for Word Representation
    • 14,464
    • Open Access
    SQuAD: 100, 000+ Questions for Machine Comprehension of Text
    • 2,125
    • Highly Influential
    • Open Access
    Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
    • 9,394
    • Open Access
    Teaching Machines to Read and Comprehend
    • 1,630
    • Open Access
    Bidirectional Attention Flow for Machine Comprehension
    • 1,131
    • Highly Influential
    • Open Access
    Reading Wikipedia to Answer Open-Domain Questions
    • 686
    • Highly Influential
    • Open Access
    Semantic Parsing on Freebase from Question-Answer Pairs
    • 947
    • Open Access
    The TREC-8 Question Answering Track Report
    • 829
    TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension
    • 457
    • Highly Influential
    • Open Access
    Gated Self-Matching Networks for Reading Comprehension and Question Answering
    • 421
    • Highly Influential
    • Open Access